Protein--protein interactions are of paramount and fundamental interest in biological systems. These interactions are involved in a wide variety of important biological reactions, including the assembly of enzyme subunits, in antigen-antibody reactions, in supramolecular structures of ribosomes, filaments, and viruses, in recognition and transport, in transcription regulation, and in ligand-receptor interactions. In addition, the area of protein--protein interactions has received significant attention in the area of signal transduction and biochemical pathway analysis.
Traditionally, protein--protein interactions were evaluated using biochemical techniques, including chemical cross-linking, co-immunoprecipitation and co-fractionation and -purification. Recently genetic systems have been described to detect protein--protein interactions. The first work was done in yeast systems, and was termed the "yeast two-hybrid" system. The basic system requires a protein--protein interaction in order to turn on transcription of a reporter gene. Subsequent work was done in mammalian cells. See Fields et al., Nature 340:245 (1989); Vasavada et al., PNAS USA 88:10686 (1991); Fearon et al., PNAS USA 89:7958 (1992); Dang et al., Mol. Cell. Biol. 11:954 (1991); Chien et al., PNAS USA 88:9578 (1991); and U.S. Pat. Nos. 5,283,173, 5,667,973, 5,468,614, 5,525,490, and 5,637,463.
However, while the yeast system works well, it is unsuitable for use in mammalian systems for a variety of reasons. Furthermore, the existing mammalian two-hybrid systems are neither suitable for a wide variety of cells, nor flexible, as they generally require quite highly specialized conditions. In addition, the existing mammalian two-hybrid systems are generally transient systems, rather than stable systems. Finally, these systems tend to have high background signals from non-specific interactions, giving rise to "false positives".
A number of factors make a flexible mammalian two-hybrid system highly desirable. First of all, post-translational modifications of proteins may contribute significantly to their ability to interact, yet mammalian cells have different post-translational modification systems than yeast. Thus, proteins that interact in a yeast system may not interact with the same specificity or avidity when placed in a mammalian cell. Similarly, proteins that would interact with correct post-translational processing may not be identified in a yeast system. In addition, a mammalian two-hybrid system that could be used in a wide variety of mammalian cell types would be highly desirable, since the regulation, induction, processing, etc. of specific proteins within a particular cell type can vary significantly; it would thus be a distinct advantage to assay for relevant protein--protein interactions in the relevant cell type. For example, proteins involved in a disease state could be tested in the relevant disease cells, resulting in a higher chance of identifying important protein interactions. Similarly, for testing of random proteins, assaying them under the relevant cellular conditions will give the highest chance of positive results. Furthermore, the mammalian cells can be tested under a variety of experimental conditions that may affect intracellular protein--protein interactions, such as in the presence of hormones, drugs, growth factors and cytokines, cellular and chemical stimuli, etc., that may contribute to conditions which can effect protein--protein interactions.
Thus, a robust and adaptable mammalian two-hybrid system which can work in a wide variety of mammalian cell types is highly desirable.
Accordingly, it is an object of the invention to provide compositions and methods useful in a two-hybrid system which can be utilized reproducibly and stably in mammalian cells.